Pipeline Metrics

This page shows you how to export metrics from a Kubeflow Pipelines component.For details about how to build a component, see the guide tobuilding your own component.

Kubeflow Pipelines supports the export of scalar metrics. You can write a listof metrics to a local file to describe the performance of the model. Thepipeline agent uploads the local file as your run-time metrics. You can view theuploaded metrics as a visualization in the Runs page for a particularexperiment in the Kubeflow Pipelines UI.

See the.

The metrics file has the following requirements:

• The file path must be /mlpipeline-metrics.json.
• must be a numeric value.
• format can only be , RAW, or not set.

• Click one of your experiments. The Runs page opens showing the top twometrics, where top is determined by prevalence (that is, the metrics withthe highest count) and then by metric name.The metrics appear as columns for each run. The following example shows the accuracy-score androc-auc-score metrics for two runs within an experiment:

Visualize the output of your component by writing out metadata for an outputviewer.